Re: [yang-doctors] Yangdoctors last call review of draft-ietf-lsr-ospfv3-extended-lsa-yang-14

[Acee Lindem <[e-mail]>]

Hi Mahesh, 

Thanks for the review - a lot of good comments. See inline and -16 version.  

> On Jun 15, 2023, at 5:18 PM, Mahesh Jethanandani via Datatracker <[e-mail]> wrote:
> 
> Reviewer: Mahesh Jethanandani
> Review result: On the Right Track
> 
> Document reviewed: draft-ietf-lsr-ospfv3-extended-lsa-yang
> 
> Status: On the right track
> 
> I have marked it as On the Right Track, because of some of the points discussed
> below.
> 
> Summary:
> 
> This document defines a YANG data model augmenting the IETF OSPF YANG model to
> provide support for OSPFv3 Link State Advertisement (LSA) Extensibility as
> defined in RFC 8362. OSPFv3 Extended LSAs provide extensible TLV-based LSAs for
> the base LSA types defined in RFC 5340.
> 
> Nits
> 
> Please add a section on Instructions to RFC editors stating what they should do
> with references such as RFC XXXX.
> 
> It would be nice to have some consistency between having description and
> reference statements start on a new line or on the same line as the statement.
> Right now, they are all over the place.
> 
> Some of the descriptions are very cryptic. E.g.
> 
>      leaf forwarding-address {
>        type inet:ipv4-address;
>        description
>          "Forwarding address";

I updated the ones that were brief and cryptic. Note that you almost have to have knowledge of RFC 5340 and RFC 8362 to understand the encodings. 


> 
> s/Description/description in the YANG model. Actually, I was surprised that
> pyang did not complain, but yanglint did.
> 
> libyang err : Invalid character sequence "Description", expected a keyword.
> (Line number 318.) libyang err : Parsing module "ietf-ospfv3-extended-lsa"
> failed. YANGLINT[E]: Parsing schema module
> "[e-mail]" failed.


Fixed - I’m surprised pyang didn’t complain as well. 


> 
> s/Addrss/Address/

Fixed. 


> 
> s/E-/Extended / in all descriptions.

When referring to the actual LSAs, it is should be “E-“. For example, E-Router-LSA. In other cases, it is spelled out. See RFC 8362. 


> 
> Comments:
> 
> The grouping such as ospfv3-e-lsa-as, ospfv3-e-lsa-area, ipv6-fwd-addr-sub-tlv
> etc. are used in one place only. Is there a reason why this has not been pulled
> inline where it is used? Did not check for all groupings, but if there is only
> one use of them, ideally they should be inlined.

I consolidated these for the link, area, and AS scoped LSDBs. I left the fowarding-address Sub-TLV in its own grouping consistent with the other Sub-TLVs. 



> 
> No need to repeat parent name in the child. Just length will do in the
> following. See Section 4.3.1 of RFC 8407. E.g.
> 
>    container route-tag-sub-tlv {
>      description
>        "Route Tag Sub-TLV";
>      leaf route-tag-sub-tlv-length {

Fixed.


> 
> Why a double -- in  container unknown--tlv {?

Fixed.

> 
> A pyang compilation of the model with —ietf and —lint option was clean.
> 
> There are no examples of configuration instance data in the draft. It would be
> helpful not only to validate the model, but also help folks who want to use the
> model.

There are only two booleans that are config=true. We can look at this though.

> 
> A idnits run of the draft reveals a few issues. Please address them.
> 
>   Checking boilerplate required by RFC 5378 and the IETF Trust (see
>  https://trustee.ietf.org/license-info):
>  ---------------------------------------------------------------------
> 
>     No issues found here.
> 
>  Checking nits according to
>  https://www.ietf.org/id-info/1id-guidelines.txt:
>  ---------------------------------------------------------------------
> 
>     No issues found here.
> 
>  Checking nits according to https://www.ietf.org/id-info/checklist :
>  ---------------------------------------------------------------------
> 
>     No issues found here.
> 
>  Miscellaneous warnings:
>  ---------------------------------------------------------------------
> 
>  == The copyright year in the IETF Trust and authors Copyright Line
>     does not match the current year
> 
>  == Line 1266 has weird spacing: '... allows  a rou...'
> 
>  -- The document date (October 17, 2019) is 1337 days in the past.
>     Is this intentional?
> 
>  Checking references for intended status: Proposed Standard
>  ---------------------------------------------------------------------
> 
>     (See RFCs 3967 and 4897 for information about using normative
>     references to lower-maturity documents in RFCs)
> 
>  == Outdated reference: draft-ietf-bfd-yang has been published as RFC
>     9127
> 
>  ** Downref: Normative reference to an Experimental RFC: RFC 1765
> 
>  ** Downref: Normative reference to an Experimental RFC: RFC 4973
> 
>  ** Downref: Normative reference to an Informational RFC: RFC 5309
> 
>  ** Downref: Normative reference to an Informational RFC: RFC 5714
> 
>  ** Downref: Normative reference to an Informational RFC: RFC 6987


These idnits are fixed.


Thanks,
Acee


> 
>     Summary: 5 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment
>     (--).
> 
> 
>